Computer Science Enrollment Falls for the First Time in 20 Years

⏱️ Lectura: 8 min

Computer science enrollment in the United States fell for the first time in 20 years, according to a Stanford University economist cited by Yahoo Finance. The data marks a sharp turn after nearly two decades of almost uninterrupted growth in enrollment for this major.

📑 En este artículo

1. TL;DR
2. What happened
3. Context and history
   1. Three decades of CS enrollment
4. Study details: what the data says (and doesn’t say)
5. How to verify the trend at your university
6. Impact and analysis
7. What’s next
8. Frequently Asked Questions
   1. Which university or economist reported the drop?
   2. Is this the first time computer science enrollment has dropped in 20 years?
   3. Is ChatGPT directly causing the drop?
   4. How can I check the trend at my university?
   5. Does this mean it’s no longer worth studying programming?
   6. What do tech companies say about junior hiring?
9. References

The drop comes right as ChatGPT and other AI coding assistants became everyday tools for junior and senior developers. For anyone closely following programming and AI, the question is direct: is generative AI changing the decision of thousands of students to study software?

TL;DR

• Computer science enrollment fell for the first time in 20 years in the U.S., according to a Stanford economist.
• The report links the drop to the mass adoption of ChatGPT, launched by OpenAI in November 2022.
• The decline breaks nearly two decades of sustained growth in CS enrollment, which began after the dot-com crash of 2003-2005.
• The CRA’s Taulbee Survey measures computer science enrollment year over year at U.S. universities.
• The data coincides with reports of lower junior hiring at tech companies that already use AI coding assistants.
• The figure can be verified with public IPEDS data, the U.S. NCES database.
• The phenomenon reopens the debate over whether programming remains a profitable career in the face of AI.

What happened

According to Yahoo Finance, a Stanford economist noted that computer science enrollment dropped for the first time in two decades at American universities. The report doesn’t focus on a single campus: it describes an aggregate trend that breaks a growth pattern that had held since the mid-2000s.

The timing isn’t a coincidence. The drop falls within the same window in which ChatGPT went from a novelty to a daily tool for coding, debugging, and automating junior tasks. That doesn’t prove causation on its own, but it does place the phenomenon within a very specific context: the rise of generative AI in software work.

Context and history

Computer science enrollment hasn’t always grown. After the dot-com bubble burst, between 2003 and 2005, enrollment fell sharply: thousands of dot-com companies closed and the tech job market contracted. From there, with the rise of giants like Google, Amazon, and Facebook and increasingly high salaries, the major saw nearly twenty years of sustained growth.

That upward cycle coincided with the rise of coding bootcamps, the explosive growth of GitHub as a portfolio platform, and the narrative that programming was the most direct path to a high salary without needing a traditional hard-science degree. Computer science became the fastest-growing major, relatively speaking, at many public universities in the U.S.

Three decades of CS enrollment

Period	CS enrollment trend	Main driver
2003-2005	Decline after the dot-com bubble burst	Dot-com company closures and tech unemployment
2006-2022	Sustained growth for nearly two decades	Rise of Big Tech, high salaries, software demand
2023-2026	First drop in 20 years, according to the cited report	Adoption of AI assistants like ChatGPT and Copilot for junior tasks

Study details: what the data says (and doesn’t say)

The report cited by Yahoo Finance attributes the drop to a Stanford economist, but doesn’t itself publish the exact decline figure or a breakdown by university. That matters: before drawing big conclusions, it’s worth separating the confirmed fact (enrollment dropped for the first time in 20 years) from the interpretation (that ChatGPT is the sole cause).

⚠️ Heads up: correlation isn’t causation. The drop in computer science enrollment coincides in time with the adoption of ChatGPT, but also with other factors like rising college tuition costs and tech hiring cycles in recent years.

What can be measured independently is year-over-year enrollment volume. In the United States, that figure doesn’t depend on a single news report: there are two public sources that track it systematically, and any reader can check them.

How to verify the trend at your university

IPEDS (Integrated Postsecondary Education Data System) is the first source, the NCES system run by the U.S. Department of Education that reports enrollment and degrees by field of study, institution by institution, every year. The second is the Computing Research Association’s (CRA) Taulbee Survey, specific to computer science departments.

A minimal example for exploring public higher education data with Python, using the Department of Education’s College Scorecard API:

import requests

API_KEY = "YOUR_API_KEY"
url = "https://api.data.gov/ed/collegescorecard/v1/schools"
params = {
    "api_key": API_KEY,
    "fields": "school.name,latest.academics.program_percentage.computer",
    "per_page": 20
}

resp = requests.get(url, params=params)
data = resp.json()
for school in data["results"]:
    print(school["school.name"], school.get("latest.academics.program_percentage.computer"))

This script prints, per institution, the percentage of students in computing-related programs. It’s a starting point for comparing different years and seeing whether the reported drop shows up at specific universities.

The second block isn’t about statistics, it’s about the cause being discussed: how far automation of junior tasks has come. A typical example of what a coding agent solves in seconds today:

$ claude "write a Python function that validates an email address with regex and add tests with pytest"

That kind of task, common in a first job or a programming internship, is exactly the work that a few years ago required hours from a junior developer. It’s the piece that supports the hypothesis that AI is changing the calculus for thousands of future students.

Impact and analysis

If the drop is confirmed with official IPEDS or Taulbee Survey data in coming cycles, the impact isn’t just academic. Fewer computer science students today means fewer engineers graduating in four or five years, right when the industry still doesn’t know how much demand for human talent it will keep having against increasingly capable coding agents.

flowchart TD
    A["ChatGPT launches (Nov 2022)"] --> B["Coding assistants automate junior tasks"]
    B --> C["Companies hire fewer interns and junior devs"]
    C --> D["Students reconsider studying CS"]
    D --> E[("Computer science enrollment")]

The diagram summarizes the hypothesis connecting the dots in the report: it’s not that AI directly empties classrooms, but that it changes the job market signals that future students use to decide their career. If landing that first junior job becomes harder, the major loses some of its relative appeal against other options.

💡 Tip: if you work at a university or CS department, your institution’s annual Common Data Set usually publishes declared enrollment by major and is faster to check than IPEDS for a single campus.

What’s next

The next hard data point that would confirm or nuance this report will come with the publication of the current academic cycle’s Taulbee Survey and IPEDS’ annual update. That will show whether the drop is widespread nationally or concentrated in certain types of universities.

It’s also worth following junior hiring reports from Big Tech, since they’re the other half of the story: if entry-level hiring recovers, the narrative that “programming no longer pays off” quickly loses steam.

📖 Summary on Telegram: See summary

Try it yourself: go to the IPEDS website and look up the enrollment-by-field-of-study report for your university to see if the national trend holds in your case.

Frequently Asked Questions

Which university or economist reported the drop?

The data was attributed to a Stanford University economist and reported by Yahoo Finance, though the article doesn’t detail the full name of the original study or the exact decline figure.

Is this the first time computer science enrollment has dropped in 20 years?

According to the cited report, yes: it would be the first drop since the major began its sustained growth cycle after the dot-com crash of the early 2000s.

Is ChatGPT directly causing the drop?

The report establishes a timing coincidence, not proof of isolated causation. Other factors, like tuition costs or tech hiring cycles, may also play a role.

How can I check the trend at my university?

By checking IPEDS or your institution’s public Common Data Set, which report enrollment and degrees by field of study year over year.

Does this mean it’s no longer worth studying programming?

The report doesn’t say that. It documents a drop in enrollment, not a conclusion about the major’s future value, which depends on how demand for human talent evolves alongside AI.

What do tech companies say about junior hiring?

The article doesn’t include its own hiring figures; for that, it’s worth checking reports from the companies themselves or separate labor data firms.

References

• Yahoo Finance: original report on the drop in computer science enrollment citing a Stanford economist.
• IPEDS (NCES): public system from the U.S. Department of Education with enrollment and degree data by field of study.
• Computing Research Association (CRA): organization that publishes the Taulbee Survey, the annual reference on enrollment in computer science departments.
• Wikipedia: ChatGPT: context on ChatGPT’s launch in November 2022 and its subsequent adoption.

📱 Like this content? Follow @programacion on Telegram for daily tech content in Spanish: quick summaries, fresh content every day.

Imagen destacada: Foto de Vitaly Gariev en Unsplash